Crime Quarterly No. 5 SA CRIME QUARTERLY No 5 SEPTEMBER 2003 29 V ictim surveys are commonly used to establish the nature and extent of crime in particular areas. The survey findings are usually used concurrently with police crime statistics, since they complement each other. Police crime statistics provide a measure of offences reported to and recorded by the police, and are influenced by changes in reporting behaviour as well as recording rules and practices. Victim surveys measure both reported and unreported crimes and are not affected by changes in reporting and police recording rules. However, victim surveys are based on estimates from a sample of the population, and can therefore be subject to sampling error and other methodological limitations.1 In late 2002, the Institute for Security Studies conducted a victimisation survey in the Nelson Mandela Metropolitan Municipality (NMMM). The survey is one of the studies conducted by ISS in an ongoing project with the municipality aimed at developing a crime reduction strategy for the area. The survey collected data on the extent of crime in the area, as well as general public opinion about crime, policing, safety measures used for protection against crime, and views of municipal service delivery. This article will analyse the findings of the victim survey, as well as the police crime statistics for the Nelson Mandela police stations Methodology For the victim survey, 3,300 people over the age of eighteen years were interviewed in their homes, using a structured questionnaire. The whole target area was stratified according to police station boundaries, using 1996 census data to sample the NMMM population.2 About two hundred households were approached in sixteen police station areas. In one particular police station area, one hundred households were visited, as the area is sparsely populated and houses widely dispersed. Of the 3,300 people interviewed, 42% were black, 35% white, 23% coloured and 1% Indian. The sampling of NMMM by using police station areas resulted in a slight over-representation of whites Sibusiso Masuku, Institute for Security Studies sibusiso@iss.co.za Just like most cities in South Africa, the Nelson Mandela Metro, which comprises the former Port Elizabeth, Despatch and Uitenhage municipalities, faces serious challenges in its efforts to address crime. A victim survey and analysis of police crime statistics conducted by the ISS indicate that there are high rates of property crime, violent crime and robbery in the metro. Although these are spread out, some communities are more affected by certain crime types than others. It is imperative that the municipality makes a strategic decision about areas of focus, which crime types to deal with first, and about appropriate and effective crime reduction interventions linked to service delivery. FINDING LOCAL SOLUTIONS: Crime prevention in the Nelson Mandela Metro SA CRIME QUARTERLY No 5 SEPTEMBER 2003 30 MASUKU and under-representation of coloured and black population groups. According to the Municipal Demarcation Board population estimates, blacks constitute 54%, coloureds 23%, whites 17%, and Indians 1% of the population in the area.3 The difference between the actual race profile of the area, and that of the realised survey sample can be attributed to the fact that there are more police stations servicing former white suburbs than predominantly black residential areas. For instance, an ISS study in the former Uitenhage municipality in 2000 showed that there was one police official for every 800 people in KwaNobuhle police station, whereas the Uitenhage police station had one police official for every 300 people.4 Crime levels in Nelson Mandela Metro Respondents were asked whether they had been a victim of a range of serious crimes in the Nelson Mandela Metro area over a specific time period. The data from their responses for the year 2002, which indicates crime levels in the area for that year, is reported in this article. As a standard procedure for household surveys, certain crimes are considered ‘household crimes’ while others are ‘individual crimes’. Household crime would include burglary, all forms of theft, and murder. Individual crime includes robbery, assault, car hijacking and sexual assault. Of the 3,300 people interviewed, 23% said that they had been victims of crime in 2002. The levels of crime reported to the NMMM victim survey in 2002 follows the usual trend for surveys of this sort. The most common crimes involve some type of theft and burglary, followed by violent crime aimed at property, such as robbery. Levels of interpersonal violence, such as assault and rape, were low. Of those crimes covered in the survey, the most common crime occurring in NMMM in 2002 was burglary of homes, followed by robbery and theft out of motor vehicles. Percentages for murder, sexual assault and car hijacking were not included because they were too small (Figure 1). Figure 2 shows the levels of victimisation in each of the Nelson Mandela police station areas in the past five years. According to the survey, 28% of people living in Nelson Mandela Metro have been victims of crime in the past five years. Ten of the 17 police stations in the metro experienced higher crime levels than the average: Mount Road and Swartkops police station areas recorded the highest levels of victimisation at 34%, whereas Despatch and Kinkelbos police station areas reported the lowest levels of victimisation at 18% and 10% respectively. It is interesting to note that, with the exception of Despatch and Kinkelbos, there was not much variation in the overall levels of victimisation in the police station areas. This is most likely due to high levels of property crimes reported in the survey, given that property crimes tend to affect all communities indiscriminately. Once data had been gathered on the actual levels of crime experienced in the area, respondents were asked to indicate which crimes they thought occurred most often in their areas. Their perceptions of the most recurrent crimes correspond with actual 4 4 Figure 1: Percentage of respondents who have experienced crime in NMMM Source: ISS Nelson Mandela Metro victim survey, 2002 0 Robbery 9Burglary Theft out of vehicle Assault 1 2 3 4 5 6 7 8 9 10 % 2 2 Car theft 1 Stock theft SA CRIME QUARTERLY No 5 SEPTEMBER 2003 31MASUKU crime levels as reported by victims of crime in the survey. This shows that public views on the crimes that affect them most are not unrealistic: home burglary (40%), robbery (24%) and theft (12%) were believed to be the crimes occurring most often in the metro. Rape, theft of vehicles and assault were also highlighted. Crimes recorded by the police in the NMMM The 1996-2000 police crime statistics were analysed to assess crime trends and changes in reported and recorded crimes in the Nelson Mandela Metro.5 Commonly experienced serious crime types were selected, of which most were also covered by the victimisation survey. A comparison between victim survey results and the official police crime figures is problematic because the definitions of crimes used by the police and those used in the survey questionnaire are not always the same. Comparisons are even more difficult when the dates Figure 2: Percentage of people who said they were victims of crime in each of the police station areas in NMMM, 1998 - 2002 Source: ISS Nelson Mandela Metro victim survey, 2002 0 10Kinkelbos 19Despatch 25KwaNobuhle 26Bethelsdorp 26KwaDwesi 27New Brighton 28Gelvandale 28Total 30Algoa Park 30Kamesh 30Kabega Park 30Walmer 31KwaZakhele 32Humewood 32Uitenhage 32Motherwell 34Swartkops 34Mount Road 5 10 15 20 25 30 35 % of victims of the survey and police crime statistics do not correlate. Therefore this is not intended as a comparison between the NMMM victim survey results and the official police crime figures, but rather an analysis and interpretation of available data at a given time, in order to provide some context for the survey results. Figure 3 shows three categories of the most serious crime types recorded by the police, namely property crime, violent crime and robbery. Property crime includes all forms of burglary, motor vehicle crime, stock theft, miscellaneous theft, shoplifting, arson and malicious damage to property. Violent crime includes murder, attempted murder, rape, and all forms of assault, while robbery includes robbery with aggravating circumstances and other robberies. Police crime statistics support the findings of the victim survey. The SAPS recorded crimes show that victim survey, and concerns about the willingness of respondents to disclose information about such crimes, the estimates of crimes such as assault and sexual assault are not considered reliable – whether in the form of police records or the victim survey. The police crime figures for the year 2000 were used to compare the rates of recorded crime in each of the police stations in the NMMM (figure 4). Most police stations in the wealthier suburbs recorded high rates of property crime and robbery and relatively low rates of violent crime. The police stations in the underdeveloped suburbs registered high rates in all crime categories, with slightly more violent crimes and robberies than property crimes. Notably, KwaZakhele, New Brighton and Motherwell police stations recorded the highest rates of robbery. They were followed by KwaNobuhle, KwaDwesi, Humewood, Kamesh, Algoepark and Swartkops. The high rate of robbery reported in Nelson Mandela Metro police stations is disturbing. Robbery is an indiscriminate crime and anyone can be a victim. This crime type impacts negatively on the public image of the area, and often increases public fear of crime. Therefore crime reduction strategies should focus on robbery, and interventions dealing with robbery and other crimes should be developed. Risk of victimisation Both the victim survey and police data show that crime levels differ across the Nelson Mandela Metropolitan area. Opportunities for crime are enhanced or reduced by the attractiveness of the target, levels of security, lifestyle patterns, or a combination of these factors. Therefore certain people and certain areas are more vulnerable to particular crime types at a particular time. SA CRIME QUARTERLY No 5 SEPTEMBER 2003 32 MASUKU property crime is the most reported crime in the Nelson Mandela metro. The reporting of property crime stabilised between 1996 and 1997, and began to increase from 1998 to 2000. On the other hand, violent crime decreased slightly between 1997 and 1998, and slowly began to increase in 1999 and 2000. Robbery, on the other hand, has been steadily increasing since 1997. Robberies are the biggest threat for people leaving in the Nelson Mandela Metro. Other robberies, including muggings, increased by 37% between 1996 and 2000, and by 16% between 1999 and 2000. Aggravated robbery, including all forms of armed robbery, increased by 33% between 1996 and 2000. Also increasing at an alarming rate was vehicle theft, home burglary and common assault. Stock theft, attempted murder and murder showed a substantial decrease between 1996 and 2000. Police crime figures show a substantial reporting of violent crime, particularly assault and rape that do not feature prominently in victim surveys. This is not surprising, since victimisation surveys tend to be weak in measuring interpersonal violent crimes such as assaults and sexual assaults. The victims of these crimes often feel uncomfortable discussing their experience of these crimes with strangers. Due to the small number of incidents reported to the Figure 3: Number of most serious crimes recorded by the police in the NMMM, 1996 - 2000 Source: SAPS Crime Information Analysis Centre 0 1996 39,740 23,145 4,030 1997 39,530 22,281 3,783 1998 43,573 20,521 4,433 1999 45,663 22,116 5,365 2000 46,659 23,618 6,223 5,000 10,000 15,000 20,000 25,000 30,000 35,000 40,000 45,000 50,000 Property crime Violent crime Robberies SA CRIME QUARTERLY No 5 SEPTEMBER 2003 33MASUKU Figure 5 shows how the different race groups experience the risk of burglary reported in the victim survey. Indians (13%), blacks (10%) and whites (9%) were more at risk than coloureds (6%). Whites and Indians also reported higher levels of theft and theft out of vehicles. These groups are targeted in respect of property crime because they are seen to own more valuable consumer goods, whereas blacks may be targeted as a result of poor security in their areas. ‘Soft’ targets and inadequate surveillance systems create opportunities for victimisation. The strategy often used by property crime criminals in selecting their targets is to go for easily accessible and unguarded property. They assess how easily a residence or car may be entered, and how well the site is protected.6 On the other hand, the risk of violent crime category shows some interesting results. According to the victim survey, black and coloured population Figure 4: Percentage of violent crime, property crime and robbery reported, 2000 Source: SAPS Crime Information Analysis Centre 0 84511 8489 81415 75717 73325 71129 61633 59437 55639 51841 45748 441244 43849 421444 371746 Mount Road Humewood Uitenhage Swartkops Kabega Park Kinkelbos Walmer Algoa Park Gelvandale Despatch Bethelsdorp KwaDwesi Kamesh Motherwell KwaNobuhle New Brighton KwaZakhele 20 40 60 80 100 % Figure 5: Risk of burglary by race of victim, 2002 Source: ISS Nelson Mandela Metro victim survey, 2002. 0 Indians 2 4 6 8 10 12 14 13 Black 10 White 9 Coloured 6 Violent crime Robberies Property crime 67528 66727 SA CRIME QUARTERLY No 5 SEPTEMBER 2003 34 MASUKU groups are more at risk of both robberies and assaults than whites. While blacks were more at risk of robbery than coloureds, coloured people living in the NMMM were more likely to be victims of assault than blacks (figure 6). More than half (57%) of people who were victims of robbery said they were robbed on the streets in the area where they live. Twenty-three per cent said they were robbed in their home environment and seventeen per cent said they were robbed near shops. The victims of assault also reported that they were victimised on the streets in their neighbourhood (32%) or in their home environment (31%). Most victims of these assaults or robberies were between 28 and 59 years of age. People in this age group are a perfect target for robbery because they are more likely to be employed and own valuable consumer goods. Reporting of crime The majority (94%) of the families who reported a murder in the victim survey indicated that the case was also reported to the police, whereas about 60% of assault victims and 48% of robbery victims said that they reported these incidents to the police. The high murder report rate is not surprising since all murders or death in any form need to be registered for certification. The reporting of assaults to the police was also unusually high; however, it remains unsatisfactory given that most victims of assault know the perpetrators. On the other hand reporting of street robberies is usually low. Often victims do not report these crimes because they are perceived to be petty, and most people know that the chances of recovering their property are slim. Victims of street robberies also have little hope that perpetrators will be apprehended. People who do report street robberies are usually unable to give clear descriptions of the perpetrators, which hampers the effectiveness of police in apprehending the suspects.7 The reporting of car theft (86%), burglary (70%) and theft out of vehicles (67%) to the police was fairly high. This may well be due to the need for a case number when claiming insurance. On the other hand, reporting of stock theft to the police was very low, perhaps because in urban areas stock theft is perceived as a petty crime. As the victims of property crimes mostly do not know the offenders, the apprehension and prosecution rates for these crimes also tend to be low. Conclusion Understanding localised crime trends and patterns is crucial in developing an effective crime reduction strategy. Just as many other cities and towns in South Africa, the Nelson Mandela Metro is faced with serious crime challenges. Although crime seems to be spread throughout the metropolitan area, some neighbourhoods are affected more than others by particular types of crime. Wealthy suburbs are experiencing high rates of property crime and in some cases robbery, while relatively poor areas experience high rates of property crime and violent crime. While competent policing and an effective criminal justice system are major requirements for crime control, more interventions are needed. Crime, particularly in poor areas, is closely linked to various socio-economic problems associated with high levels of poverty, low standards of education, a lack of opportunity for employment and self- development, and a lack of effective social services. Figure 6: Risk of robbery and assault by race of victim, 2002 Source: ISS Nelson Mandela Metro victim survey, 2002 0 Robbery 1 2 3 4 5 6 7 6 4 2 Assault 4 6 2 Blac k Coloured White SA CRIME QUARTERLY No 5 SEPTEMBER 2003 35MASUKU Therefore effective and sustainable crime reduction will require a major focus on social development programmes. Municipalities have a mandate to coordinate and provide most of the needed social services that can impact on crime rates. However, because of the limited available resources in some areas, and the vast size of the Nelson Mandela Metro, difficult choices will need to be made about which parts of the municipality to prioritise in the fight against crime. The data above shows that areas with major social development problems also have high crime rates, which suggests that these areas should be prioritised. Central business areas should also be a priority because of their economically strategic position. It will be impossible to address all types of crime at the same time, however; it would therefore make sense to select three or four priority crimes for immediate attention. Since violent crime seems to be a major problem in the Nelson Mandela Metro, crime reduction projects specifically aimed at reducing violence should be implemented in areas experiencing high levels of violent crime. The business of selecting crime types and prioritising areas is highly political, and will require extensive consultation. It would make perfect sense for the Nelson Mandela Metro Municipality to navigate and co-ordinate this delicate process. Endnotes 1 Home Office Statistical Bulletin, The 2001 British Crime Survey. 2 1996 census data was the latest national census data available at the time of the survey. 3 Municipal Demarcation Board. www.Demarcation.org.za/Munic_Profiles/New_layout/ Inndex.html 4 S. Masuku, T. Maepa and E. Pelser, Towards a crime reduction strategy in Uitenhage, Unpublished ISS report, 2000. 5 Year 2000 was the last year that full year police station crime statistics were available. 6 P. Cromwell, In their Words: Criminals on crime: an anthology, second edition, Roxbury publishing company, Los Angeles, 1999. 7 Ibid.